Paul Richards liltingly intones:
>
> In reply to Ben Laurie who said
> >
> > I don't understand why LINGER should be used. According to my man, SO_LINGER
> > causes the close() to block until all data is sent. It should have no effect
> > on the sending of the data. So, what's it all about?
> >
>
> I thought that's exactly what the problem was, some buggy SVR4 networking
> code doesn't send data sitting in the kernel part of the networking code if
> the application closes the socket. Setting SO_LINGER would work around
> that bug. This is my understanding from memory though, someone should confirm
> this.
>
This is indeed the case, at least for UnixWarez.
chuck
Chuck Murcko N2K Inc. Wayne PA chuck@telebase.com
And now, on a lighter note:
Bugs, pl. n.:
Small living things that small living boys throw on small
living girls.
|